Constructor for the DataManager class. Subclassed DataManagers will likely need to override this constructor. If you aren't manually

Build network for binocular downstream task It takes an extra argument head, that is called with the features and a dictionary img

Add camera parameters to the observation dictionnary produced by Habitat-Sim In-place modifications.

Determines which colormap to use based on set colormap type Args: outputs: the output tensors for which to apply colormaps on

A helper function for _apply_fn_to_fields, applying a function to all fields of dict_ Args: dict_: The dictionary to apply the fu

Broadcasts all tensors in a dictionary according to batch_shape Args: dict_: The dictionary to broadcast. Returns:

Calculate the maximum image height that can be rendered in the time budget Args: camera_object: the camera object to use for rend

Helper to print out any warnings regarding the way the viewer/loggers are enabled

compute curvature loss

compute point-wise approximation of curvature loss used in HashSDF

project 3d bbox to each frame to produce per-frame coarse fg masks. Theses masks are useful for weighted ray sampling in foreground and backgr

compute proposal network loss for bg sampler

sample additional ray samples from the proposal samples and compute the eikonal loss. this aims to regularize the internal geometry of the rec

compute foreground point cloud regularization loss use point-wise threshold by default.

compute plane point cloud regularization loss

compute point cloud regularization loss on the plane sdf field

Auxiliary function of radial_and_tangential_undistort() that computes residuals and jacobians. Adapted from MultiNeRF: https://github.com/goog

This function: - first downsizes the image with LANCZOS inteprolation, which is better than bilinear interpolation in

utility tool to move the cursor up on the terminal Args: x: amount of lines to move cursor upward